Vehicle Routing

Vehicle Routing Problem (VRP) with capacity, time windows, and driver constraints

Previewproblem type vrpmanifest 2.0

Objectives#

What the solver optimizes. Enabled objectives are weighed together.

Objective	What it means	Direction	On by default	Weight
Minimize Total Time `minimize_time`	Minimize aggregate route time across all vehicles (travel + service)		Yes	2.0
Minimize Total Distance `minimize_distance`	Minimize the total distance traveled by all vehicles		Yes	1.0
Minimize Vehicles Used `minimize_vehicles`	Minimize the number of vehicles required	minimize	Yes	0.7

The setup values you send with a problem. The console renders them as form fields.

Input	Name	Type	Required	Default	Description
`n_customers`	Number of Customers	number	Yes	50	Delivery/pickup locations to visit
`n_vehicles`	Number of Vehicles	number	Yes	5	Available vehicles for routing

Input	Name	Type	Required	Default	Description
`vehicle_capacity`	Vehicle Capacity	number	No	100	Maximum load capacity per vehicle

Input	Name	Type	Required	Default	Description
`depot_lat`	Depot Latitude	number	No	33.749	Latitude of the depot/starting location
`depot_lon`	Depot Longitude	number	No	-84.388	Longitude of the depot/starting location
`radius_miles`	Radius (miles)	number	No	50	Radius around depot for customer generation
`distance_metric`	Distance Metric	select	No	haversine	How distances between coordinates are measured. 'haversine' reads [lat, lon] in km; 'euclidean' reads planar [x, y]; 'euclidean_rounded' rounds each leg to an integer, the convention of published capacitated-routing benchmarks. With a planar metric, speed_kmh is coordinate units per hour.

Input	Name	Type	Required	Default	Description
`demand_distribution`	Demand Distribution	select	No	uniform	How customer demand is generated

Input	Name	Type	Required	Default	Description
`use_osrm`	Use Road Network (OSRM)	boolean	No	True	Use real road distances and speed limits instead of straight-line estimates. Requires OSRM server.
`speed_kmh`	Average Speed (km/h)	number	No	32.19	Fallback average vehicle speed when OSRM is not available. Default is 20 mph city driving (32.19 km/h).
`service_time`	Service Time (minutes)	number	No	3	Default service time per customer stop when not using per-customer dataset values.
`service_time_per_customer`	Use Per-Customer Service Time	boolean	No	True	Use each customer's service_time value from dataset instead of the default.
`vrp_solver`	VRP Solver	select	No	auto	Routing engine (sent as options.engine). 'auto' picks the best engine installed where the solve runs and names it in the result.
`polish_tnc`	TNC Polish	boolean	No		Apply 7-operator TNC steepest descent after main solver for extra refinement.

A hard constraint must hold or the problem is infeasible; a soft one adds a penalty when broken.

Constraint	What it requires	Kind	On by default
Vehicle Capacity `vehicle_capacity`	Each vehicle cannot exceed its load capacity	hard	Yes
Pickup & Delivery Capacity `pickup_delivery_capacity`	Handle both pickup (adds to load) and delivery (removes from load)	hard	No
Maximum Driver Hours `max_driver_hours`	Maximum driving/working hours per route	hard	Yes
Customer Time Windows `time_windows`	Customers must be visited within specified time windows	hard	No
Driver Breaks `driver_breaks`	Mandatory break requirements during routes	soft	No
Visit All Customers `visit_all_customers`	All customers must be visited exactly once	hard	Yes
Return to Depot `return_to_depot`	Vehicles must return to depot after completing route	hard	Yes
Minimize Vehicles Used `minimize_vehicles_constraint`	Use as few vehicles as possible	soft	Yes
Vehicle Types `vehicle_types`	Use different vehicle types with varying capacities and speeds	hard	No
Route Balance `route_balance`	Balance workload across drivers	soft	No
Driver-Customer Affinity `driver_affinity`	Prefer assigning drivers to customers they've served before	soft	No

Checked before a solve, so an impossible problem is refused with the reason.

Check	What it tests	On failure
Fleet capacity	Total stop demand must fit inside n_vehicles x vehicle capacity.	fail
Single-stop capacity	No single stop may demand more than one vehicle carries.	fail
Time-window reachability	Each stop's window must be reachable on a direct trip from the depot, and the vehicle must be able to return before the depot closes.	fail
Shift length	Each stop's out-and-back trip (drive + service) must fit inside the longest vehicle shift.	fail
Fleet shift hours	Total service time must fit inside the fleet's summed shift hours.	fail

Named starting sets of constraints.

Preset	Use it for	Constraints
Basic VRP `basic`	Simple capacity-constrained routing	`vehicle_capacity`, `visit_all_customers`, `return_to_depot`, `minimize_vehicles_constraint`
Time-Constrained VRP `time_constrained`	Routing with time windows and driver hour limits	`vehicle_capacity`, `max_driver_hours`, `time_windows`, `visit_all_customers`, `return_to_depot`
Commercial Delivery `commercial_delivery`	Full-featured routing for commercial delivery operations	`vehicle_capacity`, `max_driver_hours`, `time_windows`, `driver_breaks`, `visit_all_customers`, `return_to_depot`, `minimize_vehicles_constraint`, `driver_affinity`